Paris, Facundo N. (2018) Deep learning para la predicción de la viscosidad en un microvicosímetro capilar. / Deep learning to predict yhe viscosity in a capilar microviscometer. Master in Physical Sciences, Universidad Nacional de Cuyo, Instituto Balseiro.
![[img]](/style/images/fileicons/application_pdf.png)
| PDF (Tesis) Spanish 7Mb |
Abstract in Spanish
Se desarrolló una red neuronal multicapa destinada a la predicción de la viscosidad de fluidos a partir de las curvas de posición vs tiempo proporcionadas por el Microviscosímetro capilar diseñado por el Dr.Morhell y el Dr.Pastoriza. La motivación que propició la red fue solucionar la problemática que se presenta en las mediciones donde la dinámica del fluido se ve alterada por fenómenos propios a las condiciones de borde del microcanal. Para ello se realizaron numerosas mediciones en fluidos con diferentes coeficientes de viscosidad y a distintas temperaturas para realizar un entrenamiento supervisado de la red neuronal de manera de obtener una mejor predicción. La red neuronal desarrollada para los primeros 5 segundos de la medición mostró una muy buena generalización, probada en mediciones de plasma sanguíneo, incluso en aquellas mediciones donde la dinámica estaba alterada.
Abstract in English
We developed a Neural Network Regression to predict uid viscosity from the position vs. time curves provided by the Microviscometer designed by Dr. Morhell and Dr.Pastoriza. The motivation for the network was to solve a problem presented in the measurements when the dynamics of the fluid were altered by the border conditions of the microchannel. We performed numerous measurements in fluids with different viscosity coefficients and at different temperatures to do a supervised training of the neural network. The neural network developed for the rst 5 seconds of the measurement showed a very good generalization, tested in blood plasma measurements, even in those measurements where the dynamics were altered.
| Item Type: | Thesis (Master in Physical Sciences) |
|---|---|
| Keywords: | Neural networks; Redes neuronales; [Machine learning; Aprendizaje profundo; Microviscomeer; Microviscosimetro; Microfluidcs; Microfluidica] |
| References: | [1] Morhell, N. Microviscosímetro capilar para fluidos complejos. 2015. [2] SI-ANALYTICS-GmbH. Si analytics capillary viscometry. URL http//:www.si-analytics.com. [3] Engineering, B. Wells-brookeld cone-plate microviscosimeter. URL www.brookfieldengineering.com. [4] Anton-Paar. Anton paar mcr rheometer series. URL http//:www.anton-paar.com. [5] Spagnolie, S. E. Complex fluids in biological systems. 2014. [6] Holsworth, R., Wright, J. Blood viscosity: The unifying parameter in cardiovascular disease risk. Holist Prim Ore, 13, 2012. [7] Rosenson, R., McCormick, A. Uretz, E. Distribution of blood viscosity values and biochemical correlates in healthy adults. tomo 42 (8), pags. 1189-1195. 1996. [8] Riopel, L., Fouron, J., Bard, H. Blood viscosity during the neonatal period: the role of plasma and red blood cell type. The Journal of Pediatrics, 100, 449-453, 1982. [9] Ramamurthy, R. S., Brans, Y. W. Neonatal polycythemia: I. criteria for diagnosis and treatment. Pediatrics, 68 (2), 168-174, 1981. [10] Dintenfass, L. Blood microrheology: viscosity factors in blood flow, ischaemia, and thrombosis. 2a ed., pags. 168-174. Appleton-Century-Crofts, 1971. [11] Casanova, M. A., Martín-Ancel, A. Policitemia en el recién nacido. An Pediatr Contin, 10 (3), 135-141, 2012. [12] Buonocore, G. A practical aproach to neonatal diseases. Springer, 2012. [13] Barnes, H. A., Hutton, J. F., Walters, K. An introduction to rheology. Elsevier, 1989. [14] Segur, J. B., Oberstar, H. E. Viscosity of glycerol and its aqueous solutions. Industrial & Engineering Chemistry, 43 (9), 2117-2120, 1951. [15] White, J. L. Principles of polymer engineering rheology. John Wiley & Sons, 1990. [16] Christopher, R. H., Middleman, S. Power-law flow through a packed tube. Industrial & Engineering Chemistry Fundamentals, 4 (4), 422-426, 1965. [17] White, F. Fluid mechanics. 1998. [18] Oertel, H. Prandtl's essential of fluid mechanics. 2a ed. 2004. [19] Saha, A. A., Mitra, S. K. Effect of dynamic contact angle in a volume of fluid (vof) model for a micro fluidic capillary flow. Journal of Colloid and Interface Science, 339 (2), 461-480, 2009. [20] Waghmare, P. R., Mitra, S. K. A comprehensive theoretical model of capillary transport in rectangular microchannels. Microfluid Nano fluid, 12, 53-63, 2012. [21] Waghmare, P. R., Mitra, S. K. On the behavior of a capillary surface in a wedge. PNAS, 63 (2), 292-299, 1969. [22] Haykin, S. Neural networks and learning machines. 3a ed. 2009. [23] Gurney, K. An introduction to neural networks. UCL Press, 1997. [24] Goodfellow, I., Bengio, Y., Courville, A. Deep learning. MIT Press, 2016. http: //www.deeplearningbook.org. [25] Ng, A. Machine learning yearning. 2018. http://www.deeplearning.ai. [26] Google. Tensor ow. 2015. https://www.tensorflow.org/. [27] Cannon instrument company. https://www.cannoninstrument.com/en. [28] Reynolds, O. Phil trans royal soc london. tomo 177. 1886. |
| Subjects: | Physics > Redes neuronales |
| Divisions: | Gcia. de área de Investigación y aplicaciones no nucleares > Gcia. de Física > Materia condensada > Bajas temperaturas |
| ID Code: | 758 |
| Deposited By: | Tamara Cárcamo |
| Deposited On: | 07 Oct 2019 13:51 |
| Last Modified: | 07 Oct 2019 13:51 |
Repository Staff Only: item control page
